Nucleic acid amplification provides a means for making more copies of a nucleic acid sequence that is relatively rare or unknown, for identifying the source of nucleic acids, or for making sufficient nucleic acid to provide a readily detectable amount. Amplification is useful in many applications, for example, in diagnostics, drug development, forensic investigations, environmental analysis, and food testing. Many methods for amplifying nucleic acid sequences in vitro are known, including polymerase chain reaction (PCR), ligase chain reaction (LCR), replicase-mediated amplification, strand-displacement amplification (SDA), “rolling circle” types of amplification, and various transcription associated amplification methods. These known methods use different techniques to make amplified sequences, which usually are detected by using a variety of methods. PCR amplification uses a DNA polymerase, oligonucleotide primers, and thermal cycling to synthesize multiple copies of both strands of a double-stranded DNA (dsDNA) or dsDNA made from a cDNA (U.S. Pat. Nos. 4,683,195, 4,683,202, and 4,800,159). LCR amplification uses an excess of two complementary pairs of single-stranded probes that hybridize to contiguous target sequences and are ligated to form fused probes complementary to the original target, which allows the fused probes to serve as a template for further fusions in multiple cycles of hybridization, ligation, and denaturation (U.S. Pat. No. 5,516,663 and EP 0320308 B1). Replicase-mediated amplification uses a self-replicating RNA sequence attached to the analyte sequence and a replicase, such as Q.beta.-replicase, to synthesize copies of the self-replicating sequence specific for the chosen replicase, such as a Q.beta. viral sequence (U.S. Pat. No. 4,786,600). The amplified sequence is detected as a substitute or reporter molecule for the analyte sequence. SDA uses a primer that contains a recognition site for a restriction endonuclease which allows the endonuclease to nick one strand of a hemimodified dsDNA that includes the target sequence, followed by a series of primer extension and strand displacement steps (U.S. Pat. Nos. 5,422,252 and 5,547,861). Rolling circle types of amplification rely on a circular or concatenated nucleic acid structure that serves as a template used to enzymatically replicate multiple single-stranded copies from the template (e.g., U.S. Pat. Nos. 5,714,320 and 5,834,252). Transcription associated amplification refers to methods that amplify a sequence by producing multiple transcripts from a nucleic acid template. Such methods generally use one or more oligonucleotides, of which one provides a promoter sequence, and enzymes with RNA polymerase and DNA polymerase activities to make a functional promoter sequence near the target sequence and then transcribe the target sequence from the promoter (e.g., U.S. Pat. Nos. 5,399,491, 5,554,516, 5,130,238, 5,437,990, 4,868,105 and 5,124,246, PCT Pub. WO 1988/010315 A1 and US Pub. 2006-0046265 A1). Nucleic acid amplification methods may amplify a specific target sequence (e.g., a gene sequence), a group of related target sequences, or a surrogate sequence, which may be referred to as a tag or reporter sequence that is amplified and detected in place of the analyte sequence. The surrogate sequence is only amplified if the analyte target sequence is present at some point during the reaction. Modified nucleic acid amplification methods may amplify more than one potential target sequence by using “universal” primer(s) or universal priming. One form of PCR amplification uses universal primers that bind to conserved sequences to amplify related sequences in a PCR reaction (Okamoto et al., 1992, J. Gen. Virol. 73(Pt. 3):673-9, Persing et al, 1992, J. Clin. Microbiol. 30(8):2097-103). Methods that use universal primers often are paired with use of a species-specific, gene-specific or type-specific primer or primers to generate an amplified sequence that is unique to a species, genetic variant, or viral type, which may be identified by sequencing or detecting some other characteristic of the amplified nucleic acid. Anchored PCR is another modified PCR method that uses a universal primer or an “adapter” primer to amplify a sequence that is only partially known. Anchored PCR introduces an “adaptor” or “universal” sequence into a cDNA and then uses a primer that binds to the introduced sequence in subsequent amplification steps. Generally, anchored-PCR uses a primer directed to a known sequence to make a cDNA, adds a known sequence (e.g., poly-G) to the cDNA or uses a common sequence in the cDNA (e.g., poly-T), and performs PCR by using a universal primer that binds to the added or common sequence in the cDNA and a downstream target-specific primer (Loh et al., 1989, Science 243(4888):217-20; Lin et al., 1990, Mol. Cell. Biol. 10(4):1818-21). Nested PCR may use primer(s) that contain a universal sequence unrelated to the analyte target sequence to amplify nucleic acid from unknown target sequences in a reaction (Sullivan et al, 1991, Electrophoresis 12(1):17-21; Sugimoto et al., 1991, Agric. Biol. Chem. 55(11):2687-92).
Chamberlain, et al., (Nucleic Acid Research, (1988) 16:11141 11156) first demonstrated multiplex PCR analysis for the human dystrophin gene. Multiplex reactions are accomplished by careful selection and optimization of specific primers. Developing robust, sensitive and specific multiplex reactions have demanded a number of specific design considerations and empiric optimizations. This results in long development times and compromises reaction conditions that reduce assay sensitivity. In turn, development of new multiplex diagnostic tests becomes very costly. A number of specific problems have been identified that limit multiplex reactions. Incorporating primer sets for more than one target requires careful matching of the reaction efficiencies. If one primer amplifies its target with even slightly better efficiency, amplification becomes biased toward the more efficiently amplified target resulting in inefficient amplification, varied sensitivity and possible total failure of other target genes in the multiplex reaction. This is called “preferential amplification.” Preferential amplification can sometimes be corrected by carefully matching all primer sequences to similar lengths and GC content and optimizing the primer concentrations, for example by increasing the primer concentration of the less efficient targets. Incorporation of inosine into primers in an attempt to adjust the primer amplification efficiencies (Wu, et al., U.S. Pat. No. 5,738,995 (1998)) has also been used. Another approach is to design chimeric primers, wherein each primer contains a 3′ region complementary to sequence-specific target recognition and a 5′ region made up of a universal sequence. Using the universal sequence primer permits the amplification efficiencies of the different targets to be normalized. See, Shuber, et al., Genome Research, (1995) 5:488 493; and U.S. Pat. No. 5,882,856. Chimeric primers have also been utilized to multiplex isothermal strand displacement amplification (U.S. Pat. Nos. 5,422,252, 5,624,825, and 5,736,365). Since multiple primer sets are present in multiplex amplification reactions, multiplexing is frequently complicated by artifacts resulting from cross-reactivity of the primers. All possible combinations must be analyzed so that as the number of targets increases this becomes extremely complex and severely limits primer selection. Even carefully designed primer combinations often produce spurious products that result in either false negative or false positive results. The reaction kinetics and efficiency is altered when more than one reaction is occurring simultaneously. Each multiplexed reaction for each different specimen type must be optimized for MgCl.sub.2 concentration and ratio to the deoxynucleotide concentration, KCl concentration, amplification enzyme concentration, and amplification reaction times and temperatures. There is competition for the reagents in multiplex reactions so that all of the reactions plateau earlier. As a consequence, multiplexed reactions in general are less sensitive than the corresponding uniplex reaction. Another consideration to simultaneous amplification reactions is that there must be a method for the discrimination and detection of each of the targets. The number of multiplexed targets is then further limited by the number of dye or other label moieties distinguishable within the reaction. As the number of different fluorescent moieties to be detected increases, so does the complexity of the optical system and data analysis programs necessary for result interpretation. An approach is to hybridize the amplified multiplex products to a solid phase then detect each target. This can utilize a planar hybridization platform with a defined pattern of capture probes (U.S. Pat. No. 5,955,268), or capture onto a beadset that can be sorted by flow cytometry (U.S. Pat. No. 5,981,180). Due to the summation of all of the technical issues discussed, current technology for multiplex gene detection is costly and severely limited in the number and combinations of genes that can be analyzed. Generally, these reactions multiplex only two or three targets with a maximum of around ten targets. Isothermal amplification reactions are more complex than PCR and even more difficult to multiplex. See, Van Deursen, et al., Nucleic Acid Research, (1999) 27:e15. U.S. Pat. No. 6,605,451 discloses a two-step PCR multiplex reaction wherein a small amount of each primer pair is added into a first PCR reaction mix and a first amplification is performed to increase the amount of target nucleic acids in the reaction. The first reaction is stopped mid log phase and is then separated into second reactions each containing primer pairs for one of the target nucleic acids. A full amplification is then performed. Though a limited amount of each of the multiplex primer pairs is present in the first reaction, the above discussed problems common to multiplexing are still present. Further, these various primer pair species can all transfer into the secondary amplification reactions, causing common multiplex problems there as well. There is still a need, therefore, for a method, which permits multiplexing of large numbers of targets without extensive design and optimization constraints, and which avoids problems common to multiplexing in the presence of a plurality of different amplifications oligomer pairs. There is also a further need for a method of detecting a significantly larger number of gene targets from a small quantity of initial target nucleic acid.